Crossovers direct the accurate segregation of meiotic chromosomes. Several decades of effort has led to an understanding of many aspects of the meiotic recombination pathway, especially in S. cerevisiae. Although the early steps in meiotic recombination appear to be universal, there is clear variation in the processes that actually generate crossovers. A thorough understanding of meiotic recombination will therefore require studies in multiple model systems. This proposal takes advantage of features of Drosophila melanogaster that make certain experiments possible that are not possible in other model organisms. We will map the structures of recombination intermediates and products at high resolution by preserving heteroduplex DNA with dense marker spacing. We will also analyze both products of reciprocal crossovers to map gene conversion tracts. To better understand crossover control, we will test hypotheses for functions of a set of genes that encode MCM-like proteins essential for crossover formation, using a combination of genetic, biochemical, and cytological approaches. Finally, we will also study the late stages of the recombination process. Having identified a major Holliday junction resolvase complex, we will now ask whether other putative resolvases play a role in generating meiotic crossovers. These experiments will illuminate previously unknown features of meiotic recombination and further our understanding of the degree to which different species have selected alternative branches of DNA repair pathways to effect meiotic recombination.